Discrimination of Germline EGFR T790M Mutations in Plasma Cell-Free DNA Allows Study of Prevalence Across 31,414 Cancer Patients

Abstract

Purpose: Plasma cell-free DNA (cfDNA) analysis is increasingly used clinically for cancer genotyping, but may lead to incidental identification of germline-risk alleles. We studied EGFR T790M mutations in non-small cell lung cancer (NSCLC) toward the aim of discriminating germline and cancer-derived variants within cfDNA.Experimental Design: Patients with EGFR-mutant NSCLC, some with known germline EGFR T790M, underwent plasma genotyping. Separately, deidentified genomic data and buffy coat specimens from a clinical plasma next-generation sequencing (NGS) laboratory were reviewed and tested.Results: In patients with germline T790M mutations, the T790M allelic fraction (AF) in cfDNA approximates 50%, higher than that of EGFR driver mutations. Review of plasma NGS results reveals three groups of variants: a low-AF tumor group, a heterozygous group (∼50% AF), and a homozygous group (∼100% AF). As the EGFR driver mutation AF increases, the distribution of the heterozygous group changes, suggesting increased copy number variation from increased tumor content. Excluding cases with high copy number variation, mutations can be differentiated into somatic variants and incidentally identified germline variants. We then developed a bioinformatic algorithm to distinguish germline and somatic mutations; blinded validation in 21 cases confirmed a 100% positive predictive value for predicting germline T790M. Querying a database of 31,414 patients with plasma NGS, we identified 48 with germline T790M, 43 with nonsquamous NSCLC (P < 0.0001).Conclusions: With appropriate bioinformatics, plasma genotyping can accurately predict the presence of incidentally detected germline risk alleles. This finding in patients indicates a need for genetic counseling and confirmatory germline testing. Clin Cancer Res; 23(23); 7351-9. ©2017 AACR.

Figure 1. Germline and tumor-derived EGFR mutations within plasma cell-free DNA

(A) Across ddPCR results for 85 patients, germline T790M mutations (green) are present at a similar concentration as somatic T790M mutations (gray), but at a higher allelic fraction (AF). (B) Studying ddPCR results for 4 patients on treatment, each represented by a different color, the concentrations of somatic EGFR mutations decrease while the concentration of germline EGFR T790M remains constant. (C) Across plasma NGS results for 950 cases, the AF distribution for EGFR T790M (green) includes a somatic peak also seen with EGFR driver mutations (blue), as well as a heterozygous peak (arrow) also seen more clearly with a common SNP (EGFR Q787Q, gold).

Figure 2. Discriminating germline and somatic variants in plasma cell-free DNA

(A) Pre-treatment and on-treatment plasma specimens from three initial cases, all positive for germline EGFR T790M, were studied using plasma NGS. Studying all coding and non-coding variants detected, three groups of variants are evident, corresponding to the expected AF of homozygous, heterozygous, and tumor-derived variants. Variants in the tumor-derived group respond on therapy while variants in the homozygous and heterozygous groups remain at a relatively constant AF. (B) An additional 102 cases, for a total of 105 cases, were then studied using plasma NGS. Studying all coding and non-coding variants detected across 105 cases, a trimodal distribution is seen with peaks near 0% (likely tumor-derived), 49% (likely heterozygous), and 100% (likely homozygous). (C) For missense and nonsense variants, there is enrichment at low AF (arrows), where tumor-derived variants would be expected to be found. In contrast, synonymous variants, likely reflecting benign germline polymorphisms, are enriched around 50% and 100% AF.

Figure 3. Increased tumor content is associated with increased copy number variation within heterozygous group of variants

(A) AF of all variants found on plasma NGS of 105 cases positive for EGFR mutations, in increasing order of EGFR driver mutation AF (blue), with a common EGFR SNP shown (gold). (B) Studying the variant AFs between 25% and 75%, the standard deviation and absolute difference between case and population mean were calculated; both increased with an increase in EGFR driver AF.

Figure 4. Distinguishing heterozygous and tumor-derived coding variants in cases with low copy number variation

In outlier cases with high copy number variation (A), it is difficult to distinguish germline and somatic variants. When there is lower copy number variation (B), it is possible to visually distinguish which cases of germline EGFR T790M (green) are likely germline.

Figure 5. Prevalence of germline T790M

(A) Querying a database of 31,414 unique cancer patients with plasma NGS results, 48 (0.15%) were found to carry a germline EGFR T790M mutation. Non-squamous non-small cell lung cancer (NSCLC) was the dominant diagnosis in these patients. (B) As compared to the population prevalence of germline EGFR T790M in a reference cohort (0.008%), there is a higher prevalence in subjects with nonsquamous NSCLC (0.34%) but not in subjects with other cancers (0.03%, p = 0.06), suggesting germline EGFR T790M is a risk variant for lung cancer.